類別變量¶

https://pandas.pydata.org/docs/reference/api/pandas.Categorical.html

In [2]:

import numpy as np
import pandas as pd

import numpy as np import pandas as pd

In [21]:

import random

import random

In [28]:

np.random.seed(987)
data = ['A', 'B', 'C']*2
np.random.shuffle(data)

np.random.seed(987) data = ['A', 'B', 'C']*2 np.random.shuffle(data)

In [29]:

data

data

Out[29]:

['B', 'A', 'C', 'C', 'B', 'A']

In [30]:

cat = pd.Categorical(data)

cat = pd.Categorical(data)

In [31]:

cat

cat

Out[31]:

['B', 'A', 'C', 'C', 'B', 'A']
Categories (3, object): ['A', 'B', 'C']

In [32]:

cat.categories

cat.categories

Out[32]:

Index(['A', 'B', 'C'], dtype='object')

In [33]:

cat.codes

cat.codes

Out[33]:

array([1, 0, 2, 2, 1, 0], dtype=int8)

In [39]:

cat = pd.Categorical(data, ordered=True)

cat = pd.Categorical(data, ordered=True)

In [40]:

cat

cat

Out[40]:

['B', 'A', 'C', 'C', 'B', 'A']
Categories (3, object): ['A' < 'B' < 'C']

In [43]:

cat = cat.rename_categories(['Bronze','Silver','Gold'])

cat = cat.rename_categories(['Bronze','Silver','Gold'])

In [44]:

cat

cat

Out[44]:

['Silver', 'Bronze', 'Gold', 'Gold', 'Silver', 'Bronze']
Categories (3, object): ['Bronze' < 'Silver' < 'Gold']